Hacker group Unit 221B confirms accuracy of Deep Instinct’s defense against zero-day (>99%) and false positive threats (

Hacker group Unit 221B confirms accuracy of Deep Instinct’s defense against zero-day (>99%) and false positive threats (
Written by insideindyhomes

The assessment shows that deep learning, the most advanced form of AI, prevents ransomware, file-based, and PowerShell attacks

Deep Instinct, the first company to apply end-to-end AI-powered deep learning to cybersecurity, today announces the validation of its threat mitigation security capabilities by the cybersecurity experts at Unit 221B. The focus of the assessment was on the integrity of the Deep Instinct Prevention Platform. This was checked by various tests for portable, unknown and specially developed attacks and executable Python files. In addition, static and dynamic network and behavior analyzes and signature detections were also tested.

“Unit 221B evaluated the Deep Instinct Prevention Platform to determine its strengths in prevention and mitigation. We loved the ease of use both in installing the agents and managing the front end. Our team consists of professional skeptics and hackers who are only stopped by a few security protocols and prevention systems. We were happy and impressed that we failed in this case,” said Lance James, CEO and Founder of Unit 221B. “It’s this thorough, real-world testing that enhances our ability to collaboratively create solutions to the toughest cybersecurity problems.”

“Deep Instinct showed why deep learning is a game changer in combating and predicting tomorrow’s problems while improving today’s security posture,” continued Lance James.

After an in-depth analysis of the tests and associated results, Unit 221B determined that the Deep Instinct Prevention Platform during the two-month trial period all unknown and custom executables. This demonstrated the platform’s ability to block malicious files and successfully prevent malicious code execution – even if the platform had never been exposed to them before.

Key findings include:

  • 99.78% Accuracy – Deep Instinct demonstrated a combined accuracy rate of 99.78% for detecting and preventing unknown and custom attacks. Unit 221B tested Deep Instinct with an appropriate configuration suitable for networks of established customers with high existing security levels.
  • 100% of unknown attacks – Deep Instinct was successful in automatically preventing 100% of unknown attacks and 96.4% of Unit 221B’s custom attacks.
  • 60% fewer recorded alarms – SIEM/DER solutions with Deep Instinct installed recorded 60% fewer events/alerts than Microsoft Defender alone. This results in less strain on staff and reduces alert fatigue (also known as “Alert Fatigue”). This allows employees to focus more on strategic and critical tasks such as patching and system hardening.

Deep Instinct prevented 100% of the following zero-day and unknown threats:

  • Any unknown malicious transmittable file with an authentic code-signing certificate.
  • PoshC2 generated shellcode – An indication that the shellcode solution can recognize commonly generated shellcode patterns.
  • Various malicious document types including Microsoft Word, Excel and OneNote.
  • All custom malicious documents, links, HTAs, VBS and other Active Script, PowerShell and other file types – including signed and obfuscated attacks. These were created specifically for these tests and used techniques commonly used to bypass EDR (Endpoint Detection and Response Detection) detections.
  • All individual ransomware programs using various delivery methods. The product also demonstrated the ability to detect and prevent zero-day ransomware attacks.
  • Execution of each tested PowerShell script. The variety of scripts tested represents a diverse catalog of PowerShell attack tactics.

The Deep Instinct Prevention Platform prevents a wide range of malware variants before they are executed and thus proved to be effective in all tests. In addition, dangerous attacks are consistently prevented before the user even knows that he is the target of an attack. Zero-day vulnerabilities such as Vulnerabilities exploited in the infamous Kaseya attack, for example, were protected with more than 96% accuracy, as assessed by Unit 221B. No updates are necessary for all these functions.

“Unit 221B findings confirm that a deep learning approach to cybersecurity is a critical advance. As a result, the industry is finally able to get ahead of our opponents,” said Guy Caspi, CEO of Deep Instinct. “We have grown tremendously as a company over the past year and have seen great success and customer engagement with both our agent-based endpoint offering and our agentless product portfolio. Our customers choose Deep Instinct for our unique approach to stopping zero-day attacks before they happen. While most organizations focus on detection, we aim to prevent an attack before it gains access to your organization’s infrastructure, because in many cases detection is too little and too late. I am grateful to the Unit 221B team for their work on this evaluation and assessment.”

For full results and more information on Unit 221B’s Deep Instinct Prevention Platform product assessment, please visit:

#Hacker #group #Unit #221B #confirms #accuracy #Deep #Instincts #defense #zeroday #false #positive #threats

About the author


Leave a Comment